From Sunshine to Snowfall: Understanding concurrent CO2 and COS exchange in a Coniferous Forest

The net ecosystem exchange (NEE) of CO₂ can be measured using the eddy covariance (EC) technique, but separating NEE into ecosystem respiration and gross primary productivity (GPP) relies on models and tracers, making it a persistent challenge. Beyond the established nighttime and daytime flux partitioning algorithms, the trace gas carbonyl sulfide (COS) shows promise as a robust proxy for constraining GPP. Unlike CO₂, which is exchanged bidirectionally by leaves and its ecosystem level exchange being influenced by soil respiration, COS generally enters leaves unidirectionally and is fully catalyzed by carbonic anhydrase. Other sources and sinks of COS within ecosystems are typically minor and negligible.

Initial laboratory studies have determined the leaf relative uptake rate (LRU) – the ratio of COS to CO₂ deposition velocities (LRU = (FCOS/χCOS)/(GPP/χCO₂)) – to be relatively stable around 1.7 under optimal conditions. By knowing the LRU and measuring COS fluxes alongside CO₂ and COS ambient mixing ratios, GPP can be calculated. However, most laboratory measurements have been conducted under optimal conditions and further research revealed the influence of environmental factors such as drought, vapor pressure deficit (VPD) and photosynthetically active radiation (PAR) on the LRU.

Due to the high cost and sensitivity of required instruments, few studies have examined COS fluxes at the ecosystem scale, and even fewer have performed long-term monitoring. Seasonal dynamics, particularly during winter, remain largely unexplored.

To address this gap, we conducted EC measurements of COS and CO₂ fluxes at a Pinus sylvestris-dominated coniferous forest in Austria to investigate environmental influences on COS fluxes and LRU dynamics. Sampling has been continuous since May 2021, except for a two-month gap during the winter of 2021/22. We present the influence of VPD, PAR, temperature and snowfall on COS fluxes and the LRU at the ecosystem level, based on 3.5 years of measurements.